SOCIAL RESEARCH AND SOCIAL CHANGE:

MEETING THE CHALLENGE OF

GENDER INEQUALITY AND SEXUAL ABUSE
Deborah L. Rhode

Describing the relationship between social science research and social

change is a large project for limited space.1 The task reminds me of Gloria Steinems response when asked to describe the impact of the womens
movement. That felt, she said, like being asked to describe the universe
and give three examples.2 So that will be my strategy here. Let me offer
three illustrations of how contemporary social science research has transformed understandings of social problems and inspired efforts at social
reform.
Sexual harassment, acquaintance rape, and domestic violence have always been with us, but only in the last several decades have we begun to
chart their dimensions and dynamics and to craft meaningful responses.
Making gender inequality central to research on these issues has dramatically altered the content and context of policy debates. Yet our progress
remains partial, and our studies to date have been far more successful in describing the problems than in identifying solutions. I will conclude, then,
by exploring some of the barriers to essential policy-oriented research, both
at the individual and institutional level and by suggesting some strategies
for reform.
I.
Events like this one inevitably invite personal reection on the notso-good old days, and my memories are probably not atypical. When I
rst encountered sexual harassment, as a Yale student in the 1970s, it had
neither a label nor a remedy. We spoke of the experience as having a problem with a professor and the problem was always ours, never his. There
were no policies, procedures, or complaint channels that might have identied that there was indeed a problem. But into the void marched a series of
books and articles, as well as lawsuits, that identied sexual harassment
Ernest W. McFarland Professor of Law, Stanford Law School, and Director, Stanford
Center on Ethics. I am grateful to Mariko Hirose for her excellent research assistance.
1 This Essay grows out of an address presented in a plenary session entitled Transgressing Sex Segregation: The Law, Social Science, and Social Policy at the August 11,
2006, annual meeting of the American Sociological Association.
2 Deborah L. Rhode, The No-Problem Problem: Feminist Challenges and Cultural
Change, 100 Yale L.J. 1731, 1732 (1991).

12

Harvard Journal of Law & Gender

[Vol. 30

as a form of sexual discrimination and one that should be actionable under a newly passed statute, Title VII of the national Civil Rights Act.3
The initial reception in many quarters was less than enthusiastic. The
men who dominated the upper levels of employment, judicial, and policy
circles were often skeptical that sexual overtures and workplace banter
were signicant problems and that women were, or should be, offended. As
one male manager put it in a 1980s survey by the Harvard Business Review, Ive never been harassed but I would welcome the opportunity.4 For
those who did not ask for it, Phyllis Schlay reassured Congress, harassment was not a problem except in the rarest of cases.5 Many federal
judges agreed. In their view, the civil rights law was not meant to be a clean
language act6 or a remedy for the petty slights suffered by the hypersensitive.7 Only sustained, malicious, and brutal abuse should qualify for
relief.8 One trial judge expressed common views with uncommon candor:
So we will have to hear [your complaint], but the court doesnt think too
much of it.9
Research has played a major role in challenging and changing these
attitudes. Studies from a wide range of employment and educational settings have brought home the frequency and costs of harassment. Most have
found that between one-third and four-fths of women, and up to one-fth
of men, experience such abuse.10 Contrary to critics concerns, underreport3

Civil Rights Act of 1964, 42 U.S.C. 2000e-12000e-17 (2000). Among the most
inuential works was Catharine A. MacKinnon, Sexual Harassment of Working
Women: A Case of Sex Discrimination (1979). For an overview of the history of sex
harassment law, see Directions in Sexual Harassment Law (Catharine A. MacKinnon
& Reva B. Siegel eds., 2004).
4
Eliza G. C. Collins & Timothy B. Blodgett, Sexual Harassment . . . Some See It . . .
Some Wont, 59 Harv. Bus. Rev., Mar.Apr. 1981, at 76, 92.
5
Nancy E. McGlen & Karen OConnor, Womens Rights: The Struggle for
Equality in the Nineteenth and Twentieth Centuries 186 (1983).
6
Katz v. Dole, 709 F.2d 251, 256 (4th Cir. 1983) (holding that a former federal air
trafc controller had nevertheless satised the requirements for proving sexual harassment
under Title VII by showing that she suffered sustained serious harassment).
7 Zabkowicz v. West Bend Co., 589 F. Supp. 780, 784 (E.D. Wis. 1984) (holding that
warehouse worker had proven harassment occurred and that her employer did not take
effective action to x the problem).
8
Id. at 784.
9
Henson v. Dundee, 682 F.2d 897, 900 n.2 (11th Cir. 1982) (holding that plaintiffs
claim that she was denied the opportunity to attend police academy because of her refusal
to have sexual relations with her supervisor was erroneous, but nding that plaintiff had
made a prima facie showing of the elements necessary to establish a Title VII violation on
other claims).
10
For a study nding that 44% of women and 19% of men reported having experienced some form of unwanted sexual attention at work in the past two years, and nding
that other studies estimate the percentage of women suffering sexual harassment at work to
be as high as 40% to 80%, see Theresa M. Beiner, Gender Myths v. Working Realities: Using Social Science to Reformulate Sexual Harassment Law 1 (2005). For a
study nding that as many as one in two women may encounter workplace sexual harassment, see Heather Antechol & Deborah Cobb-Clark, The Changing Nature of EmploymentRelated Sexual Harassment: Evidence from the U.S. Federal Government 19781994, 57
Indus. & Lab. Rel. Rev. 443, 443 (2004). For a discussion of harassment on college cam-

2007]

Social Research and Social Change

13

ing, not overreaction, has been the common response. Only ve to twelve
percent of those experiencing harassment make any complaints, and fewer
still can afford the nancial and psychological costs of litigation.11 The
price of harassment, both for individuals and institutions, remains substantial. Many of these employees suffer signicant economic and psychological injuries including dismissals, demotions, depression, and other stress
and mental health disorders.12 Employers lose millions from decreased
productivity, increased turnover, and legal expenses.13 And everyone suffers
from the job segregation and subordination that exclude workers from positions they deserve.
In short, we have made considerable progress in analyzing the problem, but we are still a considerable distance from preventing it. Harassment
remains pervasive and formal complaints nearly doubled between 1992 and
2001.14 The prevailing assumption is that the answer lies with policies,
complaint channels, and training programs, and that these structures will
normally insulate employers from liability.15 Yet the evidence available
casts doubt on the adequacy of such initiatives. For the vast majority of victims, the costs of complaining outweigh the benets: barriers include fears
of reprisal, blacklisting, and loss of privacy, as well as doubts about the
value of responses.16 Although employers are spending an estimated $10 billion annually on training programs,17 and four states now require them for

puses and in elementary and secondary schools, see Deborah L. Rhode, Sex in Schools:
Whos Minding the Adults?, in Sexual Harassment of Working Women, supra note 3,
at 29096.
11
Johanna Grossman, The Culture of Compliance: The Final Triumph of Form Over
Substance in Sexual Harassment Law, 26 Harv. Womens L.J. 3, 2324 (2003).
12 Deborah L. Rhode, Speaking of Sex: The Denial of Gender Inequality 101
(1997).
13
Id.
14
Raymond F. Gregory, Unwelcome and Unlawful: Sexual Harassment in the
American Workplace 9 (2004).
15
For an analysis of this presumption of insulation from liability, see Susan BisomRapp, Fixing Watches With Sledgehammers: The Questionable Embrace of Employee Sexual Harassment Training by the Legal Profession, 24 U. Ark. Little Rock L. Rev. 147,
16162 (2001); Kimberly D. Krawiec, Cosmetic Compliance and the Failure of Negotiated
Governance, 81 Wash. U. L.Q. 487, 508, 540 (2003); Margaret S. Stockdale et al., Coming
to Terms With Zero Tolerance Sexual Harassment Policies, 4 J. Forensic Psychol. Prac.
65, 7172 (2003).
16
See, e.g., Gregory, supra note 14, at 134; Stockdale et al., supra note 15, at 73. See
generally Grossman, supra note 11. For an argument that combating sexual harassment
through assertiveness training for women is ineffective and that cognitive-behavioral psychology may provide insight into victims responses, see Louise Fitzgerald et al., Why Didnt
She Just Report Him?: The Psychological and Legal Implications of Womens Responses to
Sexual Harassment, 51 J. Soc. Issues 117 (1995). For advocacy of an approach to sexual
harassment law that takes its lead from the contradictory realities of womens lives and that
makes room for social science research, see Beth A. Quinn, The Paradox of Complaining:
Law, Humor and Harassment in the Everyday Work World, 25 Law & Soc. Inquiry 1151
(2000).
17 For these cost estimates, see Stuart Silverstein, Fear of Lawsuits Spurs the Birth of a
New Industry, L.A. Times, June 27, 1998, at A1.

14

Harvard Journal of Law & Gender

[Vol. 30

specied private sector workers,18 no systematic research documents the

effectiveness of these programs in producing sustained attitude changes or in
altering workplace behavior. What few studies have been done yield inconclusive results.19 Some nd no long-term improvements in attitudes and
others indicate that poorly designed programs can actually entrench gender stereotypes, encourage male backlash, and perpetuate the very biases
that they are designed to combat.20
Recent experience with Californias mandatory training law illustrates the risks. The programs that I have reviewed both highlight overly
obvious or overly trivial examples of harassment and exaggerate the risk
of personal liability for unintended offenses. According to some courses,
statements such as you look nice in that dress or acts of chivalry like holding a door open fall into a dangerous grey area.21 Even inadvertent offenders can reportedly risk losing not only their reputation but also their
home[s], car[s] and life savings.22 For the clueless in California, some
training includes instruction on what forms of workplace hugging are permissible. Although frontal hugs apparently are acceptable, side hugs or
squeezes may be objectionable. But then again, maybe not.23 Too many
individuals may end up seeing these required programs, and the legal doctrine that they describe, as overblown reactions to oversensitive feminists
who should get a life, not a law. As one exasperated male supervisor put
it in his evaluation, This appears to be a course designed by idiots for
idiots.24

18 California, Connecticut, and Maine require sexual harassment training for some private employers, and Michigan mandates its civil rights department to offer training programs
to employers, including private employers. Global Compliance, State Requirements
for Harassment Training, available at http://www.globalcompliance.com/pdf/staterequirements-for-harassment-training.pdf (last visited Dec. 3, 2006); see also Brightline
Compliance, Preventing Workplace Harassment: Sexual Harassment Training by State, http://
www.brightlinecompliance.com/training/sexual-harassment-training.html (last visited Dec.
3, 2006).
19
For a discussion of these studies, see Bisom-Rapp, Fixing Watches, supra note 15, at
163; Susan Bisom-Rapp, An Ounce of Prevention Is a Poor Substitute for a Pound of Cure:
Confronting the Developing Jurisprudence of Education and Prevention in Employment
Discrimination Law, 22 Berkeley J. Emp. & Lab. L. 1, 3036 (2001).
20
These studies are summarized in several books and articles. See Barbara Gutek, Sexual Harassment Policy Initiatives, in Sexual Harassment: Theory, Research, and Treatment 185 (William ODonohue ed., 1997); Bisom-Rapp, Fixing Watches, supra note 15, at
163; Bisom-Rapp, An Ounce of Prevention, supra note 19, at 3738.
21
The example comes from materials for an in-house training of staff at the Natural
Resources Defense Council (NRDC) in San Francisco in 2005 (on le with author).
22
This example comes from the training program required for Stanford University supervisors in 20052006 (on le with author).
23 NRDC materials, supra note 21. The trainer for the NRDC cautioned that some employees may object to any form of touching. Interview with Ralph C. Cavanagh, Attorney,
NRDC, in Stanford, Cal. (Nov. 2005).
24
This was the response of one Stanford University supervisor contained in his evaluation of the Stanford training program referenced in supra note 22 (on le with the author).

2007]

Social Research and Social Change

15

II.
A similar story can be told about our progressand its limits
concerning acquaintance rape. Once again, when I was in college, forced
sex was a not uncommon part of the campus social scene. But to borrow
the title of an early celebrated study, I Never Called It Rape.25 Mainly,
my female classmates and I never spoke of it at all, and when we did, our
euphemisms were along the lines of bad sex and a real mistake. We
held ourselves responsible, as did the culture generally, and in that setting
of shame and blame, reporting the experience to any authority was generally
unthinkable. If a woman wore short skirts or tight jeans, consumed too much
alcohol, or led a man on, what could one expect? As one judge summarized
prevailing attitudes, when a woman goes to a mans room, she certainly
[has] to realize that they [are] not going upstairs to play Scrabble.26 Any
normal man was entitled to expect something more.
Much, of course, depends on ones denition of normal and ones
denition of rape. Over the last several decades, a growing body of
research has documented the frequency and consequences of this long
unchallenged problem, as well as the attitudes that perpetuate it. An estimated one out of six American women has experienced an attempted or
completed rape,27 and about seventy percent of these incidents involve an
acquaintance.28 Studies of college age men have found that substantial numbers, typically around half, report that they would force a woman to have
sex if they could get away with it.29 A majority admit engaging in some
sexually coercive behaviors, and ten to fteen percent acknowledge having forced sex on a date.30 Yet most individuals do not consider this behavior
to be rape although it meets the legal denition.31

25

Robin Warshaw, I Never Called it Rape: The Ms. Report on Recognizing,

Fighting, and Surviving Date and Acquaintance Rape (1994).
26
State v. Rusk, 424 A.2d 720, 734 (1981) (Cole, J., dissenting).
27 For the one in six estimate, see Sexual Assault (Dean G. Kilpatrick et al. eds.), in
U.S. Dept of Justice, National Victim Assistance Academy (Anne Seymour et al.
eds., 2000), available at http://www.ojp.usdoj.gov/ovc/assist/nvaa2000/academy/J-10SA.htm.
28 For the acquaintance rape estimates, see Bureau of Justice Stat., U.S. Dept of
Justice, Crime Characteristics, available at http://www.ojp.usdoj.gov/bjs/cvict_c.htm
(last visited Dec. 3, 2006).
29 Susan Basow, Gender: Stereotypes and Roles 319 (3d ed. 1992) (thirty-ve to
sixty percent); Lynn Henderson, Rape and Responsibility, 11 L. & Phil. 127, 170 (1992)
(thirty to fty percent); Morrison Torrey, When Will We Be Believed? Rape Myths and the
Idea of a Fair Trial in Rape Prosecutions, 24 U.C. Davis L. Rev. 1013, 102324 (1991)
(close to one-half).
30
For citations of related studies, see Rhode, supra note 13, at 120; Warshaw, supra
note 25, at 1314. See also Basow, supra note 29, at 319 (Fifty percent of college men
engage in coercive and manipulative sex.).
31 See Rhode, supra note 13, at 120 (about one-half); Basow, supra note 29, at 318
(Eighty-four percent of men who admitted to committing acts that met the legal denition
of rape or attempted rape believed that their behavior denitely was not rape.).

16

Harvard Journal of Law & Gender

[Vol. 30

Part of the reason for this disconnect between legal standards and
popular perceptions is that rape by an acquaintance appears to be less injurious than rape by a stranger. In a survey of some four thousand collegeage women, over forty percent of those who acknowledged recently experiencing what the law denes as rape did not report it because they were
not sure a crime had been committed or that harm had been intended.32 Thirteen percent did not know how to make a report, and another thirty percent were deterred by the expectation of harsh or dismissive treatment in
the justice system.33 Yet many of these women, like many law enforcement
ofcials, may underestimate the injuries associated with sexual assaults
by an acquaintance. Researchers nd that these assaults are no less psychologically debilitating than rape by a stranger because they call into question a womans behavior, judgment, and sense of trust in ways that other
rapes do not.34
Social science research has been crucial not only in charting the dimensions of acquaintance rape but also in understanding the factors that
perpetuate it. These include the cultural importance of male sexual prowess as a measure of power, status, and control; the hypermasculinity fostered in all-male environments like fraternities and athletic teams; the impaired judgment resulting from substance abuse; and the assumptions that a
womans no may often mean yes, that provocative clothing is an advertisement for sex, that women often nd some coercion erotic, and so
forth.35
Yet although we have made considerable progress in identifying the
conditions that need to change, we have been far less successful in addressing them. Sexual assault policies and education programs are a standard
xture of campus life, but as with sexual harassment training, no body of
research establishes their effectiveness. To the contrary, the sparse evidence available suggests that the standard approach, which combines policies aimed at deterrence with a brief educational session, is insufcient

32

Bonnie S. Fisher, et al., U.S. Dept of Just., The Sexual Victimization of

College Women 25 (2000), available at http://www.ncjrs.gov/pdfles1/nij/182369.pdf
(respondents were permitted to choose more than one reason for not reporting the incident).
33
Id. at 2425.
34
Christine A. Gidycz & Mary P. Koss, The Effects of Acquaintance Rape on the Female Victim, in Acquaintance Rape: The Hidden Crime 270, 270 (Andrea Parrot &
Laurie Bechhofer eds., 1991); Bonnie L. Katz, The Psychological Impact of Stranger versus Nonstranger Rape on Victims Recovery, in Acquaintance Rape, supra, at 251, 251
53.
35
For a discussion of these various factors, see Katherine T. Bartlett & Deborah
L. Rhode, Gender and Law: Theory, Doctrine, Commentary 81620 (2006); Rhode,
Speaking of Sex, supra note 13, at 126; Torrey, supra note 29, at 101415. For an empirical study nding negative gender-based attitudes, pornography consumption, and alcohol
abuse to be signicant factors in predicting the likelihood of engaging in sexual violence,
see Joetta L. Carr & Karen M. VanDeusen, Risk Factors for Male Sexual Aggression on
College Campuses, 19 J. Fam. Violence 279 (2004).

2007]

Social Research and Social Change

17

to achieve lasting attitude change.36 Virtually no studies document the

inuence of these strategies on behavior. Nor do we have reason to believe that secondary education programs, most of which focus on abstinence,
deal adequately with the pressure for sexual activity that one-third of adolescents report experiencing.37
We do not lack for promising alternatives. Experts have developed a
range of curricular and policy initiatives, including more intensive and culturally sensitive education and peer counseling programs that target at-risk
groups.38 We urgently need more innovation and systematic evaluation of
these approaches. Signicant progress will require prevention programs
with some demonstrable success and the research necessary to develop
them.
III.
Similar points could be made about domestic violence. The problem
has always been with us, but until the last several decades the public had
little awareness of its dimensions or dynamics. Part of the reason is that
few statistics were available on family abuse and few cases reached the
justice system. Except in the most egregious circumstances, police, prosecutors, and judges viewed legal intervention as inappropriate or ineffectual.39 An inuential 1968 training manual for the International Association of Police Chiefs captured the prevailing wisdom concerning domestic violence calls.40 It advised police ofcers that their sole purpose was
36

For discussions casting doubt on the effectiveness of brief prevention programs, see
Patricia Frazier et al., Evaluation of a Coeducational Interactive Rape Prevention Program, 73 J. Counseling & Dev. 153, 157 (1994); Mary J. Heppner et al., Examining Immediate and Long-Term Efcacy of Rape Prevention Programming with Racially Diverse
College Men, 46 J. Counseling Psychol. 16, 17 (1999); Elena L. Klaw et al., Challenging Rape Culture: Awareness, Emotion and Action Through Campus Acquaintance Rape
Education, 28 Women & Therapy 47, 48 (2005). For research nding that most universities
take a deterrence-based approach to rape prevention, which is less effective than a public
health approach, see Roberto Hugh Potter et al., Examining Elements of Campus Sexual
Violence Policies: Is Deterrence or Health Promotion Favored?, 6 Violence Against
Women 1345 (2000).
37
For a report that nearly one-third of adolescents report having experienced the pressure to have sex, and that adolescent boys report experiencing more of such pressure than
do their female peers, see Henry J. Kaiser Family Foundation, National Survey of
Adolescents and Young Adults: Sexual Health Knowledge, Attitudes and Experiences 3 (2003). For the inadequacy of secondary sex education programs, see Michelle J. Anderson, Negotiating Sex, 78 S. Cal. L. Rev. 1401, 1434 (2005).
38
For a conclusion that intensive, sustained rape education efforts play a vital role in
dismantling rape supportive culture, see Klaw et al., supra note 36, at 48. For a description of a successful, long-term attitudinal change in some men after a multi-session intervention and a demonstration of the value of culturally relevant materials in interventions
among racial minorities, see Heppner et al., supra note 36.
39 Deborah L. Rhode, Justice and Gender 239 (1989).
40 See Sue E. Eisenberg & Patricia L. Micklow, The Assaulted Wife: Catch 22 Revisited, 3 Womens Rts. L. Rep. 138, 156 (1977) (quoting Intl Assoc. of Police Chiefs,
Training Key No. 16: Handling Domestic Disturbance Calls 94 (196869)).

18

Harvard Journal of Law & Gender

[Vol. 30

not to insure victims safety but rather to preserve the peace, and that
they should attempt to pacify [the] parties and use arrest only as a last
resort; never should they create a police problem where there [was] only a
family problem existing.41 Given such attitudes, it is perhaps not surprising that a 1982 Civil Rights Commission report found the odds of a
spousal abuse case ending up in court to be about 100 to 1.42 The prevailing assumption was that battered women were responsible for provoking
or tolerating their abuse. As so many repeatedly put it, Why dont they just
leave?43
Researchers helped supply the answer. Millions of women have had
nowhere safe to go and no way of adequately supporting themselves and
their children. Even now, the United States still has more shelters for animals than for survivors of domestic violence, and many groups, such as
gays, lesbians, immigrants, and the elderly, are especially underserved.44
Some evidence suggests that the most serious assaults and greatest risk of
homicide may occur after women attempt to leave an abusive relationship.45
As research has also made increasingly clear, the human and social
costs of unchecked domestic violence are staggering. The best estimate is
41

Id.
U.S Commn on Civil Rts., Under the Rule of Thumb: Battered Women and
the Administration of Justice 36 (1982).
43 Rhode, supra note 13, at 11419.
44 For the animal shelter claim, see Sarah M. Buel, Family Violence and the Health
Care System: Recommendations for More Effective Interventions, 35 Hous. L. Rev. 109,
111 (1998); see also Kerry Kennedy Cuomo, Speak Truth to Power, 73 Pa. B. Assn Q.
169, 170 (2002). For the difculties facing victims of same-sex domestic violence who
seek assistance, see Tara R. Pfeifer, Note, Out of the Shadows: The Positive Impact of Lawrence v. Texas on Victims of Same-Sex Domestic Violence, 109 Penn. St. L. Rev. 1251,
1251 (2005); see also Martha R. Mahoney, Legal Images of Battered Women: Redening
the Issue of Separation, 90 Mich. L. Rev. 1, 50 (1991) (Homophobia in society deters
many battered lesbians from invoking the legal system.). For an explanation of the unique
barriers facing battered immigrant women, see Mary Ann Dutton et al., Characteristics of
Help-Seeking Behaviors, Resources and Service Needs of Battered Immigrant Latinas:
Legal and Policy Implications, 7 Geo. J. on Poverty L. & Poly 245, 24748; see also
Jenny Rivera, The Availability of Domestic Violence Services for Latinas in New York
State: Phase II Investigation, 21 Buff. Pub. Int. L.J. 37, 7778 (2003) (concluding that
[l]atinas do not receive adequate intimate partner violence services. Monolingual Spanish-speaking Latinas are most adversely affected by the lack of bilingual/bicultural services, and undocumented women are the ones least likely to receive services.). For a description of the inability of most shelters to provide needed services for elderly women,
see Mary Twomey et al., Courts Responding to Domestic Violence: From Behind Closed
Doors: Shedding Light on Elder Abuse and Domestic Violence in Late Life, 6 J. Center
for Families, Child. & Cts. 73, 73 (2005) ([R]ecognition of elder abuse as a social and
legal problem is years behind child abuse and domestic violence.). For a discussion of the
inadequacy of shelters for women with disabilities, see Karen Nutter, Domestic Violence in
the Lives of Women with Disabilities: No (Accessible) Shelter from the Storm, 13 S. Cal.
Rev. L. & Womens Stud. 329, 330 (2004) ([R]esearchers . . . agree that the provisions
of protective orders and crisis shelters . . . are grossly inadequate to serve the needs of
women with disabilities.).
45 See Patricia Tjaden & Nancy Thoennes, U.S. Dept of Justice, Extent, Nature, and Consequences of Intimate Partner Violence: Findings From the National Violence Against Women Survey, 3738 (2000).
42

2007]

Social Research and Social Change

19

that about twenty-ve percent of women and eight percent of men will
experience such abuse at some point in their lifetimes.46 Domestic violence
is a leading cause of injury to women and costs the nation several billion
dollars annually in health care, absenteeism, lost wages, legal expenses,
and incarceration.47 Researchers have also documented the interplay of social, economic, and psychological factors that contribute to abuse, including
perpetrators desire for power and control, and inability to handle
anger and stress;
victims economic and psychological dependence;
inadequate societal strategies for deterrence and protection; and
family histories and cultural reinforcement of violence.48
Yet as with other issues of sexual abuse, we have a much better picture of the causes and consequences of domestic violence than the strategies that can prevent it. On some issues, the problem is not an absence of
studies but an absence of answers. For example, researchers are divided
about the effectiveness of policies requiring arrests, medical reporting of
injuries, and prosecution regardless of victim consent.49 Some studies nd
that mandatory policies are helpful in forcing the criminal justice system
to take domestic violence seriously, in deterring future abuse, and in minimizing victims vulnerability to pressure and retaliation by taking the
decision to proceed out of their hands.50 By contrast, other researchers nd
46

Id. at iii.
Bartlett & Rhode, Gender and Law, supra note 35, at 500 (estimating annual
costs of $3 billion to $5 billion); Rhode, Speaking of Sex, supra note 13, at 108 (estimating annual costs of $5 billion to $10 billion).
48 Bartlett & Rhode, supra note 35, at 46975. See generally Carolyn Puzella, Domestic Violence: Social Scientists Perspectives on the Causes of Spousal Abuse, 11 J.
Contemp. Legal Issues 37 (2000).
49 For an analysis of mandatory arrest policies as they relate to race, see generally
Miriam H. Ruttenberg, A Feminist Critique of Mandatory Arrest: An Analysis of Race and
Gender in Domestic Violence Policy, 2 Am. U. J. Gender & L. 171 (1994). For an explanation of mandatory arrest policies and their use, see Linda G. Mills, Killing Her Softly: Intimate
Abuse and the Violence of State Intervention, 113 Harv. L. Rev. 550, 562 n.55 (1999). For
a description of the impact of a Kentucky mandatory reporting law on the provision of
domestic violence services, see generally Linda K. Bledsoe et al., Understanding the Impact of Intimate Partner Violence Mandatory Reporting Law, 10 Violence Against
Women 534 (2004). For an analysis of mandatory prosecution policies, see Cheryl Hanna,
No Right to Choose: Mandated Victim Participation in Domestic Violence Prosecutions,
109 Harv. L. Rev. 1849 (1996).
50 For the support of mandatory arrest policies, see Christopher D. Maxwell et al.,
Natl Inst. of Justice, Research in Brief: The Effects of Arrest on Intimate
Partner Violence: New Evidence from the Spousal Assault Replication Program
(2001); see also Joan Zorza, Must We Stop Arresting Batterers?: Analysis and Policy Implications of New Police Domestic Violence Studies, 28 New Eng. L. Rev. 929 (1994). For
the support of mandatory reporting laws, see Bledsoe et al., supra note 49, at 538. For the
support of mandatory prosecution policies, see Hanna, supra note 49. For one of the only
studies asking battered women about their own preferences (in which approximately threefourths supported mandatory arrest and medical reporting laws and two-thirds supported
47

20

Harvard Journal of Law & Gender

[Vol. 30

little deterrent effect, increased risks of retaliation, and increased likelihood that victims will fail to call the police or seek needed services.51 Critics
also worry that disregarding victims own assessments of the risks and
benets of legal intervention further victimizes victims, by compounding
their trauma and further eroding their sense of efcacy and self-esteem.52
More large-scale research is necessary to test these competing claims
and to assess the viability of alternative, more qualied policies that make
victims safety the preeminent concern. For example, some studies suggest
that arrest and prosecution work best for perpetrators with the most to
lose from criminal sanctions, and that risks of violence escalate with those
who have the least to lose, those who are unemployed, unmarried, poorly
educated, and/or already have a criminal record.53 How best to tailor law
enforcement policy in light of different offender proles is a crucial societal question on which further research is essential.
The same is true concerning the batterer intervention programs that
serve as an alternative to incarceration or that are offered in prison. Since
the late 1970s, a growing number of jurisdictions have adopted such programs in the hopes of rehabilitating offenders through attitude change
and improved behavioral skills such as anger management.54 Despite the
mandatory prosecution policies), see Alisa Smith, Its My Decision, Isnt It?: A Research
Note on Battered Womens Perceptions of Mandatory Intervention Laws, 6 Violence
Against Women 1384, 139496 (2000).
51 For an overview of this issue, see Bartlett & Rhode, supra note 35, at 49192.
For criticisms of mandatory arrest policies, see, for example, Robert C. Davis et al., The
Deterrent Effect of Prosecuting Domestic Violence Misdemeanors, 44 Crime & Delinq.
434, 441 (1998) (We found no evidence that prosecution outcomes affected the likelihood
of recidivism in domestic violence misdemeanor cases.); Deborah Epstein et al., Transforming Aggressive Prosecution Policies: Prioritizing Victims Long-term Safety in the
Prosecution of Domestic Violence Cases, 11 Am. U. J. Gender Soc. Poly & L. 465, 467
(2003) (For many battered women, prosecution of their batterers actually creates a greater
long-term risk of harm.). For criticism of mandatory medical reporting, see Bledsoe et al.,
supra note 49, at 53940; Ariella Hyman & Ronald A. Chez, Mandatory Reporting of Domestic Violence by Health Care Providers: A Misguided Approach, 5 Womens Health
Issues 208, 209 (1995) ([M]andatory reporting can actually discourage battered women
from seeking medical care.); Michael A. Rodriguez et al., Mandatory Reporting of Intimate Partner Violence to Police: Views of Physicians in California, 89 Am. J. Pub. Health
575, 57778 (1999) (summarizing objections of medical professionals); Mia M. McFarlane,
Mandatory Reporting of Domestic Violence: An Inappropriate Response for New York
Health Care Professionals, 17 Buff. Pub. Int. L.J. 1 (1999). For criticism of mandatory
prosecution policies, see Hanna, supra note 49, at 186566; Epstein et al., supra, at 46769.
52 See Mills, supra note 49, at 56883; Epstein et al., supra note 51, at 47072.
53 For a seminal study detailing these risks, see Lawrence W. Sherman et al., The Variable Effects of Arrest on Criminal Careers: The Milwaukee Domestic Violence Experiment,
83 J. Crim. L. & Criminology 137 (1992); see also Barbara Fedders, Lobbying for Mandatory-Arrest Policies: Race, Class, and the Politics of the Battered Womens Movement,
23 N.Y.U. Rev. L. & Soc. Change 281, 281 (1997) ([R]ecent research suggests that arrest actually may be ineffective in deterring domestic violence among low-income men,
and that women of color may have reasons not shared by white women for being dubious
about tougher arrest policies.).
54
Larry Bennett & Oliver Williams, Controversies and Recent Studies of
Batterer Intervention Program Effectiveness 1 (2001), available at http://www.vawnet.
org/DomesticViolence/Research/VAWnetDocs/AR_bip.pdf.

2007]

Social Research and Social Change

21

prevalence of these initiatives, insufcient research has assessed their

effectiveness and ndings are inconclusive.55 Some studies have found no
differences in attitudes toward domestic violence and recidivism rates between men in treatment programs and those on probation.56 Other research
nds lower recidivism among men who completed the treatment program.57 However, the signicance of these results is often compromised by
high drop-out rates and non-random assignment systems that screen out
men with low motivation.58 We know even less about the long-term effectiveness of violence prevention programs aimed at adolescents.59 In the
last ve years, the federal government has spent some $3.2 billion to combat
violence against women.60 It is shameful that so few resources have gone
to learning how to target our interventions most effectively. Crafting sensible public policy on this issue will require a much more solid knowledge base than is currently available.
IV.
Sexual abuse is just one of many areas in which the nation has urgent unmet needs for research targeted at social problems. At the same
time, much of the social science scholarship that accumulates annually
vanishes without apparent inuence. About three-quarters of this published
work is never cited.61 The disconnect between the priorities of academics
and the needs of policy communities demands closer attention. What are
the institutional and individual incentive structures that get in the way of
getting the research that we need?
At the institutional level, the most obvious problem is money. Talk is
cheap; good research is not. It often satises more constituencies to spend
limited resources on efforts to x the problem than on evaluation of those
55 See Ileana Arias et al., Violence Against Women: The State of Batterer Prevention
Programs, 30 J.L. Med & Ethics 157, 161 (2002).
56
Bennett & Williams, supra note 54, at 6. For recidivism rates compared to perceptions of the potential for future recidivism by victims and batterers, see Arias et al.,
supra note 55, at 162.
57
Bennett & Williams, supra note 54, at 8, 1213; Shelly Jackson et al., Natl
Inst. of Justice, Batterer Intervention Programs: Where Do We Go from Here?
13 (2003); Varsha Pandya & Wallace J. Gingerich, Group Therapy Intervention for Male
Batterers: A Microethnographic Study, 27 Health & Soc. Work 47, 48 (2002).
58 For an experiment that screened participants, see Bennett & Williams, supra note
54, at 6 (Brooklyn Experiment). For experiments with high attrition rates, see Jackson
et al., supra note 57, at 23.
59
For an analysis of the sparse research available and recommendations for future
work, see Dean Peacock & Emily Rothman, Working with Young Men Who Batter:
Current Strategies and New Directions (2001), available at http://www.vawnet.org/
DomesticViolence/Research/VAWnetDocs/AR_juvperp.pdf.
60 Alison Siskin, Cong. Research Service, Violence Against Women Act: History, Federal Funding, and Reauthorizing Legislation 6 (2001) (updated for scal
years 20012005).
61
David P. Hamilton, Research Papers: Whos Uncited Now?, 251 Science 25, 26
(1991).

22

Harvard Journal of Law & Gender

[Vol. 30

efforts. Moreover, in many contexts, those who develop intervention programs are not eager to hear that they are ineffective, particularly if it is
not clear what would work better or how expensive that alternative would
be. For example, American employers are willing to invest vast sums in
workplace harassment programs despite a lack of evidence about their efcacy, because they help insulate the organization from liability. Managers
can claim that they are making good faith efforts to prevent abuse. That
claim would be undercut by research indicating that their programs were
not affecting workplace behavior. For similar reasons, college campus administrators may be reluctant to subsidize studies that could expose the
extent of acquaintance rape and the inadequacy of programs designed to
address it. To alter these incentive structures, it may be necessary to impose program evaluation requirements or to provide greater public subsidies for such research. Protecting the condentiality of results obtained by
institutions that undertake voluntary research into these issues might also
allay concerns that adverse ndings might create more problems than
they would solve.62
We also need to address some of the disincentives for empirical research faced by individual scholars. Much of the work on social problems in
my own eld, law, is glutted with theory but starved for facts. Many law
professors have internalized W. H. Audens celebrated injunction, Thou
shalt not . . . commit [a] social science, and with reason.63 Data are a luxury
good. Few legal academics have ready access to nancial support on the
scale necessary for major empirical research, and raising funds through
grants or private donations is a skill that few have shown interest in acquiring. Sociologists are more entrepreneurial, but many have had to
abandon valuable project designs for lack of resources.
Moreover, money is only one of many obstacles. Rigorous empirical
work generally involves a substantial commitment of time and an equally
substantial tolerance for drudgery. Particularly for younger faculty, who
face pressure to demonstrate immediate productivity, data collection is an
unattractive option. Why bother with burdensome empirical research when
you can leap instantly into print by reading others work in the comfort
of your ofce and saying what you think? In many elds, particularly in
the social sciences and professions, what yields the greatest return for scholars is not what would have the most direct value for society. The best
way of building an academic reputation is typically high theory and so-

62 For discussion of such a privilege, see Bisom-Rapp, An Ounce of Prevention, supra

note 19, at 46 n.360.
63 W.H. Auden, Under Which Lyre: A Revolutionary Tract for the Times, in Collected Shorter Poems 192757 at 221, 225 (1966). For elaboration of this point, see
Deborah L. Rhode, In Pursuit of Knowledge: Scholars, Status, and Academic
Culture 4445 (2006).

2007]

Social Research and Social Change

23

phisticated quantitative analysis, not time-consuming empirical research

that could help address practical problems.64
The obstacles to such research can be substantial even in disciplines
that claim to value it highly. C. Wright Mills, one of Americas leading
sociologists, famously confessed that I do not like to do empirical work
if I can possibly avoid it. If one has no staff, it is a great deal of trouble;
if one does employ a staff, then the staff is often even more trouble.65
Methodological complications can be equally vexing for reasons other
researchers note: computers crash; data are incomplete, irrelevant, or miscoded; [and] informants refuse to be candid or fail to return [questionnaires and] telephone calls.66 Worse still, after considerable time, expense, and effort, the results may not justify the investment. The response
rate may be too low or too unrepresentative of the sample as a whole. Alternatively, ndings may appear too obvious; they merely conrm what
everybody (especially in retrospect) already knows.67 Research on program evaluation runs the further risk of bringing unhappy tidings to administrators whose cooperation made the research possible in the rst place.
Their reaction may be to shoot the messenger, which scarcely encourages
academics to do that kind of work again.
Not only do many scholars face substantial obstacles to pursuing empirical research, but they also fail to get attention for the research that they
do undertake. Social science journals and jargon are uninviting to the uninitiated and too seldom reach the decision makers most in need of guidance. Yet writing in accessible publications for nonacademic audiences is
often viewed as dclass. Mere popularizer is an epithet to be avoided.68
For all but the celebrity scholars with access to the national media, it is
safer to produce a deeply theorized if unread tome in an obscure academic
journal than to write something useful in plain English for policy makers
or the public.
So, too, serious scholars frequently lack the time, taste, and training
for getting their research ndings to the relevant constituencies. Unless academics are afliated with a well-funded think tank, they are unlikely to
have the networks and public relations skills needed to market their work
effectively. Educating reporters can be a tedious and thankless enterprise.
They call at inconvenient moments, need time-consuming explanations,
garble quotes, or leave them on the cutting room oor. Television is particu64

Harvey Brooks, Current Criticisms of Research Universities, in The Research

University in a Time of Discontent 231, 23136 (Jonathon R. Cole et al. eds., 1994).
65
C. Wright Mills, On Intellectual Craftsmanship, in The Sociological Imagination 195, 205 (1959).
66
Peter H. Schuck, Why Dont Professors Do More Empirical Research?, 39 J. Legal
Educ. 323, 331 (1989).
67
Id.
68
John Donatich et al., The Future of the Public Intellectual: A Forum, Nation, Feb.
12, 2001, available at http://www.thenation.com/doc/20010212/forum (comments of Stephen
Carter).

24

Harvard Journal of Law & Gender

[Vol. 30

larly risky for amateurs. If my own limited experience is any guide, academics who are not seasoned performers can seldom count on looking
good and sounding good on the same occasion. A typical opportunity arises
at the last minute on a bad hair day with no time to prepare. Then, despite a
mad scramble to become presentable, the scholar gets one question on
the order of: And so professor, in the fteen seconds remaining, what
does your research tell us about how to end sexual violence?69
Yet while these obstacles to socially engaged research are substantial,
we could surely do more to help academics effectively inuence public
policy. For example, universities could target additional fellowships, stipends, workshops, and related resources to that effort. More institutions
could establish policy research centers or consortia that help to steer scholarship toward social issues where empirical work is most needed.70 More
training and support could be available to assist faculty with media outreach.
Professional associations could provide more rewards and recognition for
academics who are foot soldiers in the struggle for social change. Law
reviews could actively solicit engaged research. As we acknowledge all
that social science has helped us achieve, we must not lose sight of the
challenges that remain.

69 For a description of similar experiences, see Laurie I. Levenson, TV, or Not TV, Cal.
Lawyer, July 1997, at 95.
70 For a description of the development of such a center at Arizona State University,
see Jeffrey Brainard, A More Social Science, Chron. Higher Educ., Nov. 18, 2005, at
A22A25.